Low-voltage distribution short-circuit current peak weight compensation gray prediction method and device, terminal device and storage medium

Abstract

The embodiment of the invention discloses a low-voltage distribution short-circuit current peak weight compensation gray prediction method. The method comprises: obtaining short-circuit current original data of a low-voltage distribution system and a power supply voltage initial phase angle, a gray power prediction model optimizing the short-circuit current original data to obtain a gray power model prediction value, calculating a gray power model prediction value sequence by using the time response function, obtaining a second-order residual original data sequence through the short-circuit current original data and the gray power model prediction value, normalizing the second-order residual original data sequence to second-order residual original data x<e><(0)>(k); performing second-orderresidual gray model prediction on the second-order residual original data sequence to obtain a second-order residual prediction value [x-caret]<e><(0)>(k); obtaining a residual precision B through the second-order residual original data and the second-order residual prediction value; and weighting the second-order residual prediction value [x-caret]<e><(0)>(k) for dynamic compensation of a corrected value, and then reducing the second-order residual compensated corrected value, to obtain a short-circuit current predicted value sequence after correction.

Description

technical field [0001] The invention relates to the field of low-voltage power distribution system protection, in particular to a low-voltage power distribution short-circuit current peak value weighted compensation gray prediction method, device, terminal equipment and storage medium. Background technique [0002] With the increasing scale and capacity of the circuit system, the short-circuit capacity and short-circuit current also increase rapidly, and the short-circuit fault has become one of the most common and most harmful types of faults in the circuit system. It is of great significance to improve the safety and reliability of power grid operation by studying and extracting useful information from the fault transient process, so as to predict the short circuit and its peak value and take corresponding measures to remove the fault. At present, there are relatively few studies on short-circuit current peak prediction in low-voltage power distribution systems. Some liter...

AI Technical Summary

Problems solved by technology

[0004] 1. The general prediction model requires a large amount of raw data, takes a relatively long time, and is mostly offline calculations. It is not suitable for the online analysis with a limited number of samples required for early troubleshooting of short-circuit fault current peaks in low-voltage power distribution systems. quick forecast

[0005] 2. Gray modeling is based on the exponential law of the sequence itself. If the gray level of the data is greater, that is, the degree of dispersion is greater, the prediction accuracy will be worse.

[0006] 3. The gray prediction model is a continuous time function, which is generally higher between 2-3 points after the prediction point. As time goes by, various interference factors will intensify the impact on the system, and the gray interval of the prediction value will increase, which is not applicable medium and long term forecast

Images